AI Disclosure
Last updated 15 August 2026
Your tailored résumé is written by an AI model. It works only from material you gave us, and we check its output in code for invented facts — but it can still be wrong in ways only you can catch. Read it before you send it.
What the AI does
When you generate a tailored résumé, we send your master profile and the job posting to OpenAI's models. The model rewrites, reorders and re-emphasises your own experience against that posting. It is not writing a résumé from scratch and it is not researching you.
Two other steps use a smaller model: extracting structure from the résumé you upload, and regenerating a single bullet when you ask it to in the editor. The match score and keyword coverage are not AI — they are deterministic string matching, so the same résumé and posting always produce the same number.
What it is not allowed to do
The single biggest risk in a product like this is fabrication — a model inventing an employer, a job title, a date, or a qualification that would turn your résumé into a lie. We treat that as a bug, not as an acceptable model behaviour:
- Every company, job title and date in the output is checked in code against your source profile. Anything that does not appear there is removed before you ever see it.
- This check does not rely on instructing the model politely. It runs on the response, every time, and the product records what it had to remove.
- If a field is not in your source material, it stays empty. We do not fill gaps with plausible guesses.
Where it can still be wrong
The guard catches invented facts. It cannot catch judgement, and you should expect to correct some of these:
- Overstated emphasis. Rephrasing a contribution as more central than it was is not a fabricated fact, but it is still something you would not say in an interview.
- Dropped context. Tailoring means selecting, and it may leave out something you consider important.
- Wrong register. Vocabulary borrowed from the posting may not sound like you.
- Ordinary model error. Awkward phrasing, a garbled acronym, a bullet that does not quite parse.
This is why there is an editor
You can click any line of the preview and fix it, and what you edit is exactly what downloads. Nothing about the AI step is final. Treat the generated document as a strong first draft that you own and are responsible for — because when you send it, you are the one making the claims in it.
Your content and model training
Your résumé and the job postings you paste are sent to OpenAI to produce your result. They are not used to train models, and we do not share them with anyone else for any other purpose. See the Privacy Policy for the full list of processors.
What we do not claim
We do not claim a tailored résumé will get you an interview, will beat any particular applicant tracking system, or will be scored favourably against other candidates. We do not rank you against anyone. See the Terms of Service.